Mesoscale convective systems, which occur frequently during the warm season in midlatitudes, are significant sources of water for the agricultural areas of the Midwest. Unfortunately, numerical models often have difficulty accurately predicting the rainfall related to these systems. The rainfall, in turn, affects soil moisture which influences lower tropospheric temperature and moisture and can affect subsequent convective system formation. In recent years, several weather observing networks have been combined to form the Iowa Environmental Mesonet (IEM). The high density of surface stations in the IEM can help in better understanding mesoscale phenomena. This thesis will describe ways that the IEM data was used to better understand how small-scale variations in surface moisture and texture can affect boundary layer processes. In addition, this thesis will describe how the IEM was used to better understand sources of errors in precipitation forecasts of convective systems. For two post-precipitation events, variations in soil moisture and texture did result in modifications of the radiation balance components that likely modified air and dew point temperatures that, for the most part, were insignificant. Small areas of increased and reduced precipitation did occur as a result of the variations in soil moisture and texture, and were possibly caused by changes in low-level moisture, surface wind convergence and vertical motions. For three simulations of precipitation events, precipitation errors in the 10 km Eta model were believed to be caused by an incorrect amount of low-level wind convergence and upwards motions.